高维数据下广义线性模型自适应桥惩罚估计的变量选择

Variables Selection of Adaptive Bridge Penalty Estimation for Generalized Linear Model under High Dimensional Data

利用对数似然函数和自适应桥惩罚估计方法研究了高维数据下广义线性模型的参数估计和变量选择问题,利用对数似然函数和自适应桥方法构造惩罚估计目标函数,在适当的正则条件下,证明了自适应桥估计量的相合性和Oracle性质,通过数值模拟和实例分析验证了所提方法的有限样本性质及其优良性。

In this paper,variable selection and parameter estimation of generalized linear model with high-dimensional data is studied by using log-likelihood function and adaptive bridge penalty estimation method.By constructing penalty estimation objective function using log-likelihood function and adaptive bridge method,and under appropriate regular conditions,the consistency and Oracle properties of the adaptive bridge estimator are proved,and the finite sample properties of the proposed method are verified by numerical simulation and case analysis.